Rethinking Polarization

Jonathan Rauch

Fall 2019

In may 2016, a motorist named Cassy McWade had an accident on the interstate in western North Carolina and called her mechanic for a tow. The mechanic couldn't come, so he sent a friend from another company. In due course, Ken Shupe, of Shupee Max Towing, reached her, but as he began preparing to tow her car, he noticed a Bernie Sanders sticker on the bumper. Shupe was a Trump supporter. He told McWade he would not accept her business, suggested (by one account) that she call the government for help, and drove away.

"I'm really not interested in doing business with that clientele," he later told the local TV news station, WLOS. Asked if he thought it was fair to leave a motorist stranded, he replied, "It's not fair, but it's the norm nowadays. It's the world in which we live."

Just so: In our time, polarization has not only grown sharper but has even become its own justification. In the spring of 2018, a poll by the Pew Research Center registered yet another marker in the long series of milestones on the road to ungovernability: Democrats are now just as averse to compromise as Republicans. Only a minority in both parties (46% of Democrats, 44% of Republicans) told Pew they "like elected officials who make compromises with people they disagree with." The essence of the U.S. Constitution is to require compromise as a condition of governing. In rejecting compromise, Americans are rejecting governance. The United States and other countries have been down this road in the past, and the results are never good.

In some respects, of course, there is really not much new about polarization. It has been with us for a long time, and there is a vast political-science literature on the subject. There is also a vast journalistic  literature on it, to which I have contributed over the years. But something has changed, which requires a reconsideration of the subject.

Fifteen years ago, when polarization appeared to be a fast-emerging issue in our politics, everyone agreed on some important trends: In Congress, party-line polarization — commonly called partisanship — had risen steadily since the mid-1970s. Centrist members, always vulnerable in their swing districts, were being washed away. Gerrymandering contributed to the trend, but the same pattern prevailed in the Senate, which by definition is not gerrymandered. In Congress, party-line voting had become the rule, and there was little ideological overlap between the parties. Elite polarization was evident in both parties, albeit not symmetrically: Republicans had veered further to the right than Democrats had to the left at that point. The political class's middle was steadily hollowing out. All of those trends have grown worse in a mostly linear fashion since.

More controversial, in the 2000s, was whether polarization was a popular phenomenon as well as an elite phenomenon. Some political scientists, such as Alan Abramowitz of Emory University and Kyle Saunders of Colorado State University, marshaled evidence showing that support in the electorate for relatively centrist positions was shrinking. Another school of thought, led by Stanford's Morris Fiorina, argued that the ideological center had shrunk but not all that much, and that lots of Americans still agreed on many issues. The big change, he said, was that the parties, which once were both ideologically diverse coalitions, have sorted themselves out on ideological lines and offer increasingly extreme candidates to the voters.

Everyone agreed that both kinds of changes — in the makeup of the parties and in the ideologies of the partisans — had happened, and that both created challenges for governing, but the relative emphases mattered. If America is becoming a nation of extremists, then it will just become harder and harder to govern ourselves. If, however, the parties are becoming more ideologically distinct and coherent, that could have an upside.

There was, and still is, a case for parties with contrasting points of view, even if that sometimes translates into roadside intolerance. In 1950, a task force of the American Political Science Association declared that America's two parties were, in effect, insufficiently polarized. Called "Toward a More Responsible Two-Party System," the task force's report argued that the parties' philosophical incoherence blurred choices and accountability. In 1990, when I studied the political system in Japan, a common complaint — which I found persuasive — was that the country's ideologically piebald political factions had emptied elections of meaningful choices and therefore provided too little guidance for policy. Disagreement characterizes "the world in which we live," as the tow-truck driver said, and a healthy political system surfaces conflict instead of suppressing it.

Those were the general parameters of the conversation 15 years ago: before the Great Recession, Sarah Palin, Bernie Sanders, the Tea Party, Black Lives Matter, and of course President Donald Trump. Today, the questions that preoccupied us back then remain relevant, but in important respects they have been subsumed by another set of questions. What we mean today by polarization and what we meant then are simply not the same.


To understand how things have changed, we can begin by noticing that what Fiorina said more than a decade ago is still true: The ideological center has shrunk but has not by any means disappeared. "[O]n most issues," Fiorina wrote in 2014, "attitudes continue to cluster in the middle." And yet, the center continues to erode. In a recent analysis of American National Election Studies data, Abramowitz finds that the share of all voters placing themselves in the center of the ideological scale (or unable to place themselves) fell from 49% in 1972 to 35% in 2012. According to Pew, the share of Americans who take mixed ideological positions, rather than being consistent conservatives or liberals, stood at 39% in 2014, down 10 points from the 2004 level (which was the same as in 1994). If party sorting has slowed, it is only because both parties are too pure to be sorted much further.

Geographical sorting, meanwhile, amplifies the effects of party sorting. As of 2016, about 60% of Americans lived in so-called "landslide counties," where either Trump or Hillary Clinton received at least 60% of the major-party vote; the comparable figure was 50% in 2012 and about 40% in 1992.

Moreover, the center punches below its weight politically. The more consistently ideological and extreme an American is, the more likely he is to vote and make political contributions. Discouraged, centrists and moderates increasingly perceive elective office as a hostile environment and are reluctant to run for Congress and state legislatures — making the governing environment still more hostile to centrists and moderates, and depriving moderate voters of the opportunity to vote for compromise-minded candidates. In all of those respects, to repeat, the trends of the past few years have been more or less linear. So what is the change I allude to?

Consider the Pew figure cited above, finding that 39% of Americans take mixed ideological positions. Peek beneath the surface of that number, and you find that many of these 39% are not ideologically moderate; they are ideologically mixed up. In many cases, they hold an eclectic assortment of extreme positions.

Something analogous is also true of the two political parties. Assuredly, Democrats and Republicans are much more purely left-wing and right-wing, respectively, than they were 50 years ago. Certainly, Republicans are more hostile to government than are Democrats. Ideologically, the parties stand on opposite shores. But they still are far from internally coherent in any philosophical sense. Democrats are divided between market-oriented technocrats and self-declared democratic socialists. Republicans are in the grip of a populist takeover that defies the relatively libertarian, internationalist orthodoxy of their heretofore dominant Reaganite wing. Are the Democrats center-left or socialist? Both. Are the Republicans populist or libertarian? Again, both. We are not seeing a hardening of coherent ideological difference. We are seeing a hardening of incoherent ideological difference.

And that is really the essence of what feels like a growing division in our country. In 2017, Pew's polling found that blacks' political attitudes have not diverged significantly from whites' since 1994, or women's from men's, or college graduates' from non-college graduates'. Even across lines of age and religious observance, political attitudes have diverged only modestly. But the attitudinal gap between Democrats and Republicans has risen from 15 percentage points in 1994 to a whopping 36 points in 2017. In other words, the growing, and now gaping, divide in Americans' political values is specifically partisan. And the growth in partisanship does not reflect a clear or clean ideological divide. First and foremost, the increase in partisanship reflects, well, an increase in partisanship.

Here we reach an interesting, if somewhat surreal, question. What if, to some significant extent, the increase in partisanship is not really about anything? To put the point in a less metaphysical way, what if tribalism as such, not ideological disagreement, is behind much or even most of the rise of polarization? What if emotional identification with a partisan team is driving ideology, more than the other way around? To understand and assess this peculiar proposition, we need to expand our toolkit beyond classical political science and into social and cognitive psychology.


It increasingly appears that students of polarization 10 or 15 years ago were barking up the wrong tree. We were looking for changes in ideology when changes in feelings are more important. Several research developments have brought about that reassessment.

One of those is the growing awareness of so-called "affective polarization." This measures not differences in what partisans believe but differences in their subjective feelings toward one another. On this score, the news is pretty grim. As Pew noted in 2014,

The share of Republicans who have very unfavorable opinions of the Democratic Party has jumped from 17% to 43% in the last 20 years. Similarly, the share of Democrats with very negative opinions of the Republican Party also has more than doubled, from 16% to 38%. But these numbers tell only part of the story. Among Republicans and Democrats who have a very unfavorable impression of the other party, the vast majority say the opposing party's policies represent a threat to the nation's well-being.

Abramowitz finds that affective polarization has increased faster than issue polarization (although the two are intertwined). At least there is something both sides agree on: namely, that they can't agree. In Pew's polling, large majorities (more than three-quarters of respondents) in both parties concur that Republican and Democratic voters can't even agree on basic facts. And there is something else they agree on: In 2016, according to Pew, majorities of highly politically engaged Republicans (62%) and highly politically engaged Democrats (70%) said the other party makes them feel "afraid." When one ponders those and other such findings, one is forced to reflect that the word "hate" is too strong, but it is, alas, in the right ballpark for inter-party feeling right now. What we fear, we tend also to hate.

A second, related development is that of "negative partisanship." It's not so much that we like our own party as that we detest the other. In fact, Eric Groenendyk, of the University of Memphis, finds evidence that people hate the other party partially because they are disappointed in their own party. "[T]hey appear to be rationalizing continued identification with their party in the face of this ambivalence by reporting even more negative feelings toward the other party," he writes. "In other words, they seem to be engaging in the 'lesser of two evils' identity defense." By protecting their sense of belonging, intense partisan animosity performs what Groenendyk has called emotional rescue. The fevered view of President Obama proffered by people like Dinesh D'Souza may have been absurd, but it did serve the purpose of making every Republican leader look better by comparison. If Donald Trump is the devil incarnate, then you had better support whatever mediocre Democrat is on offer.

In any case, an implication of negative partisanship is that partisans are not so much rallying for a cause or party they believe in as banding together to fight a collective enemy — psychologically and politically a very different kind of proposition, as we see when we look at the literature on what tribalism does to the brain.

Humans are designed to be tribal. We are wired to organize ourselves socially into in-groups (our own group) and out-groups (others' groups), and to organize ourselves cognitively so that our reasoning processes and even our sensory perceptions support in-group solidarity. "Believing is belonging," as the social-psychology phrase goes — a generalization now backed by a wide-ranging and impressive research literature based on everything from controlled experiments to brain scans.

Fans of opposing sports teams perceive different events in close calls. Fans of opposing political parties perceive different facts and take different policy views depending on which party lines up on which side. Presenting people with facts that challenge an identity- or group-defining opinion does not work; instead of changing their minds, they will often reject the facts and double down on their false beliefs. This is true regardless of educational and cognitive firepower; in fact, super-smart people use their big brains to perform somersaulting rationalizations. "Extreme partisanship may be literally addictive," writes the social psychologist Jonathan Haidt. Partisans who find ways to rationalize their beliefs get a little hit of dopamine. "Like rats that cannot stop pressing a button, partisans may be simply unable to stop believing weird things."

To some considerable extent, what we are calling polarization in the United States seems not to be ideological or even rational. We may rationalize it on ideological grounds, and it may drive us toward ideological discord, but what we are in fact doing is satisfying a deep, atavistic craving to belong to an in-group and to bind ourselves to our group by feeling and displaying animosity toward an out-group. If group solidarity requires us to perform a 180-degree reversal on, say, free trade or immigration or Russia or North Korea or military action in Syria, we will flip and then rationalize the reversal. If group solidarity requires us to excuse Donald Trump for behaviors far worse than those we condemned in Bill Clinton, no problem. To paraphrase Groucho Marx, we have our principles, but if our in-group doesn't like those principles, we have other principles.

At this point, it probably goes without saying that the themes Trump ran and won on in 2016, and the themes he governs on, are short on philosophical coherence. But they have a deeper psychological coherence. Trump's appeals to ethnic and racial resentment, his portrayals of a country and culture under siege, and his populist demonization of multiple enemies offered Republicans something more appealing than any particular list of policies: They offered solidarity against a threat. On this reading, Trump may have secured his initial foothold by catering to the George Wallace portion of the party, but the remainder did not rally to Trump because they embraced his message; they embraced his message in order to rally to Trump. He offered a vivid us-versus-them story that energized one portion of the party, and then, once his followers redefined what "we" (the in-group) believe, the rest of the party preserved its identity by scrambling aboard. Although the result was to reverse Republican orthodoxy on everything from entitlement spending and trade protectionism to global alliances and the FBI, partisans felt no psychological inconsistency or lurch, because, as a result of their ideological somersaults, they continued to be aligned with the same in-group and opposed to the same out-group.

On this interpretation, the Republican base, or at least some large portion of it, likes Trump precisely because the Democratic base hates him. Polarization is not a byproduct of his policies and rhetoric; polarization is the product. His provocations and the other team's reactions satisfy partisans' craving for shared outrage against a common adversary. Within certain fairly broad boundaries, he was free to offer all sorts of policies successfully (and possibly even to shoot someone on Fifth Avenue); as long as he provoked outrage from the other side, he would elicit protective loyalty from his own side.

In that respect, Trump is not a political genius, but merely a garden-variety demagogue, pressing buttons that politicians have manipulated since Alcibiades. Talking about school choice or trade-adjustment assistance or vocational education or other ways to help the working class might be well and good, but the public's collective amygdala craved something more like WWE wrestling, an us-versus-them, good-versus-evil drama where you can stand on a chair and yell your guts out.

That does not mean that ideology plays no role in partisanship; the ideological and partisan gaps have both increased, and they are hard to disentangle (though research by the University of Maryland's Lilliana Mason finds that "the effect of issue-based ideology is less than half the size of identity-based ideology" in driving "affective" or social polarization). It also does not mean that reason is inconsequential and we are always slaves to tribalism. In fact, what follows argues the opposite. But both academic research and real-world politics over the past few years suggest that a purely political or ideological account of polarization is incomplete. We are up against a kind of tribalism here that is deeper and tougher than we had imagined. It is a powerful source of political energy, and demagogues around the world are plugging into it.


One of the most important characteristics of this "new" form of polarization is that there is nothing new about it. Tribalism has been the prevalent mode of social organization for all but approximately the most recent 2% of years that humans have lived on the planet. What needs explaining is, first, why it should be asserting itself so powerfully now after decades of relative dormancy, and, second, why our standard means of containing it seem to be failing.

As to the first question — why so much tribalism right now? — the causes are, as social scientists say, overdetermined. There is no shortage of reasons why public demand for extreme polarization might have increased. Among those commonly cited is the decline of civic organizations like unions and clubs, which has reduced individuals' sense of connectedness and agency, impelling people toward connectedness through identity and tribe. Stagnant wage growth for the less-educated causes disappointment and resentment, creating openness to demagoguery. The declining hold of organized religion and especially the collapse of mainline Protestant denominations have displaced apocalyptic and redemptive impulses into politics, where they don't belong. Identity politics on the left and market fundamentalism on the right erode the feeling of shared citizenship and identity. Changing demographics and high penetration by immigrants inspire fears of economic and cultural displacement among whites. The decline of traditionally masculine jobs and social roles leaves working-class men feeling emasculated and marginalized. The fragmenting of media isolates us in our separate information bubbles. Algorithmic social-media platforms provide a lucrative business model for viral outrage. The flowering of lifestyle diversity and consumer choice makes social differences more blatant.

It is impossible to know just how to evaluate the relative or absolute merits of those and other contributors to tribalization. Take your pick and add your own. The common theme, in any case, is that humans were designed for life in small, homogeneous groups where change was slow and choices were few. So if we find ourselves living in large, heterogeneous populations with fast-paced change and a bewildering array of choices, we may be more apt to build a tribal cocoon for ourselves: a form of emotional rescue (to use Groenendyk's term) that partisan polarization can provide if more pro-social ways of connecting fail.

That might help explain the growing demand for polarization, but what about our defenses? They are many and diverse, but a good way to think about them can be found in an important speech that Jonathan Haidt delivered at a 2017 Manhattan Institute event. Haidt argued that it is difficult for tribalistic humans to run and sustain a modern liberal society founded on compromise, toleration, and impersonal rules and institutions. Pulling it off requires getting a lot of social settings just right. Those settings include formal laws like the Constitution, informal norms like law-abidingness and truthfulness, rules-based institutions like free markets and elections, a system of education that inculcates liberal values, and public mores that honor and defend those values.

These social arrangements can restrain naked tribalism by appealing to competing values and offering us a better deal: If I master my own tribal impulses and follow liberal norms, others will do the same, and we all can enjoy more peace, prosperity, and fairness than any tribal chief or deity could provide. The Golden Rule, Locke's social-contract theory, Smith's invisible hand, Kant's categorical imperative, Mill's case for intellectual freedom, Rawls's veil of ignorance, and the U.S. Constitution's balancing of contending powers all, in their various ways, demand that we set parochialism aside and entrust our fates to abstract rules and neutral authorities, on the condition that others do the same.

Over the course of centuries, we gradually came to take these liberal settings for granted — but that was a mistake. Throughout history and throughout the world today, few societies have managed to overcome winner-take-all tribalism. And in America, we have recently taken a lot of liberties with the settings. In politics, we have shunted aside the gatekeepers and institutions that screened out sociopathic politicians before they reached the ballot. In civic life, we have succumbed to — in fact, we have valorized — cynicism about core institutions. In education, elite universities frequently encourage students to burrow into their tribal identities rather than transcend them. In media, new technologies enable and monetize outrage and extremism. In the realm of social mores, norms like forbearance, truth-telling, and respect for law have come under attack, generally for financial or political profit. These and other social settings began to drift long before Trump.

Again, there is no shortage of candidates for blame. Again, most share a common theme. Denigrating and weakening liberal values and institutions increases the fear that we cannot live with one another peaceably and thus raises the perceived level of threat. When the conservative activist Grover Norquist asserts that if more states adopt right-to-work laws, "the modern Democratic party will cease to be a competitive power in American politics," he is proposing not to come to terms with the other side but to eradicate it. The other side responds just as fearfully as one would expect. Fear is radicalizing; and, for the many entrepreneurial activists, politicians, and celebrities who cultivate fear, radicalism is personally and politically profitable.

The founders were well aware of the dangers of populism, demagoguery, and faction. They built a constitutional order designed to force compromise and impede sociopathic behavior. But the institutions they put in place to act as gatekeepers (the Electoral College, the appointed Senate) became obsolete, and the successor gatekeepers (political bosses, smoke-filled rooms, big media) came to seem undemocratic and lost their grip. Today, the road to power for a sociopath or demagogue is comparatively unobstructed. As a result, the fail-safes designed to protect the system when the settings go out of alignment have themselves begun to fail.

This is how a new form of polarization has come to be. Increasingly, partisan disagreement is rooted not in policy differences but in a sense of threat — a sense gleefully amplified by demagogues. If the other side wins, our own side's lifestyles, our values, our very ability to exist will hang by a thread. Thus the meme of the "Flight 93 election" or the death of democracy. Politics becomes a zero-sum or negative-sum conflict. Norms and constraints go out the window; every available tool and technique is weaponized. In that respect, the kind of polarization we are seeing now is not political so much as anti-political. It is certainly not what the American Political Science Association task force had in mind when it called for parties to develop more distinct and coherent policy positions.


If liberal settings like institutions, norms, trust, and social cooperation can be maladjusted, they can also be re-adjusted. We know this is so because multiple countries around the world have in fact established durable and successful liberal regimes that have withstood seemingly overwhelming challenges. We Americans have seen severe tribalism in our own past: in the early 1800s between Federalists and Democratic-Republicans, in the 1820s and 1830s between Jacksonians and Whigs, in the 1850s between the North and the South, in the civil-rights conflicts of the mid-20th century. Yet the country successfully regained equilibrium every time.

Collectively and individually, we can choose. Human individuals are in some respects wired for tribalism. But not in every respect, and not in ways we are simply helpless to resist. More important, human societies are more plastic than individuals, and they can be shaped by voluntary choices — choices in turn shaped by reason and goodwill and institutional design. Unlike our apish ancestors, modern Americans can make decisions about their lives and their polity, and those decisions will make a difference. As Haidt has said, tribalism may be pre-wired, but it is not hardwired.

Lots of people are alarmed and exhausted by the polarization industry and are fed up with it. Many are becoming aware that their buttons are being pushed and their brains are being hacked. More are resolving to do something about it by forming civic efforts with names like Better Angels, Bridge the Divide, BridgeUSA, Living Room Conversations, and the Listen First Project.

No single organization or initiative can turn polarization around. But a growing number of such efforts can make a difference in several ways. One, of course, is by reducing animosity and opening communication between individuals. Another, however, is by challenging the narrative of helplessness. Participating in the work of such groups often leaves people with a sense that there is something they can do. They feel empowered to improve the affective climate, if not in the whole country, at least in their own lives and communities. As we know from Tocqueville, when Americans set about to raise a roof or build a civic group, there is not much they cannot accomplish.

An even more important way to rebuild interpersonal connections is by strengthening institutions. Most people think little about institutions in the context of polarization. In fact, most Americans nowadays think little about institutions at all. But rebuilding institutions and thinking more institutionally are the most important pieces of the puzzle.

In three recent lectures at Princeton University, the conservative scholar and writer Yuval Levin makes a profound and compelling case for institutions. Both the public's loss of confidence in institutions and its neglect even to think about them, he argues, have left a gap in American everyday life that causes anomie and frustration and makes problems difficult to solve. "Our society is in need of something that it lacks but isn't asking for," he says. "What we are missing, although we too rarely put it this way, is not simply connectedness but a structure of social life: a way to give shape and purpose, concrete meaning and identity, to the things we do together."

By institutions he means, broadly, "'the durable forms of our associational life.' They are the forms and structures of what we do together." Their weakening makes everything else more difficult. True, society today faces many challenges, but not exceptional ones, he argues:

What does stand out about our not the strength of these pressures but the weakness of our institutions — from the family on up through the national government with lots in between — which leaves us less able to handle the pressures and to hold together. And it leaves us with something more like formless connectedness, a social life without structural supports, which can work fine for people with lots of social capital to spend but doesn't work nearly as well for those who need to build their stock of it.

Levin does not take up the problem of polarization specifically, but his point applies. In many respects, institutions are enemies of tribalism, at least in the context of a liberal society. By definition, they bring people together for joint effort on common projects, which builds community. They also socialize individuals and transmit knowledge and norms across generations. Because they are durable (or try to be), they tend to take a longer view and discourage behavior that considers only self-interest in the very short term. Because they force individuals to consult with others before making decisions and hold them accountable afterward, they filter and correct cognitive distortions. Under many conditions, they can provide stability and resources to buffer the shocks of economic and social disruption. And by organizing collective effort, they make people much more efficacious, which increases people's sense of agency and dignity. Think about how membership in the Boy Scouts, Kiwanis clubs, the American Bar Association, or the military forms us as individuals, connects us with others, builds our communities, and establishes norms that define us. In all these ways and others, human institutions were the breakthrough technology that freed Homo sapiens from our tribal chains.

Of course, institutions can act parochially, antisocially, or oppressively. They can themselves be captured by tribalism. They need oversight and healthy competition. But when we begin to treat institutions as obstacles to personal fulfillment and sweep them aside in an excess of democratizing zeal, they are replaced not by, say, wondrous social networks that connect and pacify the entire world, but by atomization, inefficaciousness, and vulnerability — which, in turn, breed fear and hostility. It is no coincidence that Donald Trump and his ilk seek to denigrate and disempower institutions at every turn. Strong institutions stand in sociopaths' way, and weak ones smooth the path for demagoguery.

Rebuilding institutions — and, just as important, noticing and valuing them — is more important for containing tribalism than pretty much anything that public policy could do. And two institutions in particular deserve strengthening: the Republican Party and the Democratic Party.

This will no doubt seem strange. Aren't the parties, almost by definition, the very font of partisan polarization? Actually, no. Paradoxically, partisanship has never been stronger, but the party organizations have never been weaker — and this is not a paradox at all. When they had the capacity to do so, party organizations engaged citizens in volunteer work, local party clubs, and social events, giving ordinary people a sense of political engagement that merely voting or writing a check cannot provide. Until they lost the power to do so, they road-tested and vetted political candidates, screening out incompetents, sociopaths, and those with no interest in governing. When they could, they used incentives like jobs, money, and protection from primary challenges to get legislators to work together and accept tough compromises. Perversely, the weakening of parties as organizations has led individuals to coalesce instead around parties as brands, turning organizational politics into identity politics.

To put the point another way, the more parties weaken as institutions, whose members are united by loyalty to their organization, the more they strengthen as tribes, whose members are united by hostility to their enemy (whether real, exaggerated, or invented). In that respect, polarization is called upon to provide solidarity when institutions cannot. If we cannot sustain institutions, we will instead create bogeymen.

In our anti-party, anti-institutional environment, none of that can easily be turned around. Still, a growing contingent of political thinkers believes that strengthening parties is possible and would be beneficial. At a minimum, rethinking the past four decades' single-minded pursuit of an ever more individualistic politics is essential.

So where does that leave us? In a swamp, but with a path out. We cannot change human nature. We are stuck with our Serengeti-evolved selves. But we are rational creatures, capable of analyzing and understanding the forces that beset us, and then capable of responding. Getting traction against affective polarization and tribalism will require some direct measures, such as civic bridge-building. Even more, it will require indirect measures, such as strengthening institutions like unions, civic clubs, political-party organizations, civics education, and others. Above all, it will require re-norming: rediscovering and recommitting to virtues like lawfulness and truthfulness and forbearance and compromise.

First, we need understanding and awareness; then we can build personal and community connections; then we can rebuild social norms and institutions. Those are the steps toward depolarization, and we have lots of history to tell us they can work. No guarantees are on offer, of course. But no new technology is needed, either. And perversely, by the sheer crudity of their attacks and the sheer baldness of their sociopathy, President Trump and his troll army may end up strengthening liberal norms and institutions by scaring us, at last, into defending them.

Jonathan Rauch is a senior fellow at the Brookings Institution. This essay is adapted from a lecture delivered at Arizona State University on February 22, 2019.


from the


A weekly newsletter with free essays from past issues of National Affairs and The Public Interest that shed light on the week's pressing issues.


to your National Affairs subscriber account.

Already a subscriber? Activate your account.


Unlimited access to intelligent essays on the nation’s affairs.

Subscribe to National Affairs.